[gmx-users] Question about gcc version

[David van der Spoel]

Justin A. Lemkul wrote:
>>> Thanks David.  I ran the test set, and got the following:
>>>
>>> $ ./gmxtest.pl all
>>> No topol.tpr file in morse. grompp failed
>>> FAILED. Check files in morse
>>> 1 out of 16 simple tests FAILED
>>> FAILED. Check files in dec+water
>>> 1 out of 14 complex tests FAILED
>>> All 63 kernel tests PASSED
>>> Error not all 45 pdb2gmx tests have been done successfully
>>> Only 38 energies in the log file
>>>
>>> For the "No topol.tpr file in morse" error, the grompp.out file shows:
>>>
>>> Fatal error:
>>> number of coordinates in coordinate file (conf.gro, 1490)
>>>              does not match topology (topol.top, 0)
>>>
>>> ...although I'm not sure why.
>> Look further up in the grompp.out file
>>
> 
> Alright, I tracked this one down to the fact that the "include" line was
> pointing to a non-existent directory (/home/gmx/top).  I changed that in the
> grompp.mdp and all is well.
> 
>>> The dec+water failure has been discussed before, although the discrepancy
>> is
>>> slightly different than that listed on the wiki:
>>>
>>> There are 5 terms to compare in the energy files
>>>
>>> LJ (SR)          step  34:      -0.21167,  step  34:    -0.171631
>>>
>>> The pdb2gmx.log file shows that there is some discrepancy in the .mdp
>>> parameters:
>>>
>>> WARNING 1 [file ../../../../em.mdp, line unknown]:
>>>   Unknown left-hand 'bd-temp' in parameter file
>>>
>>> checking input for internal consistency...
>>> processing coordinates...
>>>
>>> -------------------------------------------------------
>>> Program grompp, VERSION 3.3.3
>>> Source code file: futil.c, line: 345
>>>
>>> File input/output error:
>>> b4em.gro
>>> -------------------------------------------------------
>>>
>>> I'm guessing this is just a reflection of changes that have been made
>> between
>>> versions 3.3.2 and 3.3.3?  But I don't know what to make of the "38
>> energies in
>>> log file."  Is there some more relevant output I can post to help clarify
>> the
>>> situation?
>> Again, here you may need to look more closely in the output file. If
>> b4em.gro does not exist something went wrong before.
>>
> 
> This error seemed a bit more worrisome.  It turns out that the initial pdb2gmx
> conversion to conf.gro failed, giving a similar error to one that I reported in
> a bugzilla not too long ago (http://bugzilla.gromacs.org/show_bug.cgi?id=171). 
> This time around, pdb2gmx complained:
> 
> Fatal error:
> Atom H not found in residue 2 while adding improper
> 
> It turned out that (in the bug report) the Gromacs installation was from a
> broken Ubuntu package on that particular machine.  Looks like using gcc 4.0.1
> might result in a similarly broken package in this case.  I have installed
> gcc-3.3, used it to re-compile Gromacs 3.3.3, and all the tests pass except for
> the dec+water, which seems OK to me.
> 
If it works it works, but anyone using a Mac has gcc-4.0.x (including 
me). Mixing of packages could be an explanation.

-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++